xiv.] . WULLNER AND SALET. 195 



which would be quite right, provided hydrogen and carbon 

 could only be supposed to have one spectrum ; the impurity 

 acetylene having two. 



In 1868 Wullner, who had followed Pllicker's work from 

 1863, strongly supported his view of the existence of double 

 spectra, indicating at the same time that the difference of 

 temperature must be regarded as the sole cause of the pheno- 

 menon, adding, however, a " decomposition into further elements 

 is not to be thought of." In the case of hydrogen he showed 

 that the banded spectrum ascribed to acetylene really depended 

 upon a change in the emissive power brought about by an 

 alteration of temperature. Touching oxygen, he showed that 

 three distinct spectra may be obtained, while in nitrogen two 

 are observed. 



I may say that in my early laboratory experiments I was at 

 first led to think that, in the case of metallic vapours, Ang- 

 trom's first expressed opinion was correct, and I said so. But 

 after more experience and knowledge had been acquired I was 

 compelled by the stern logic of facts to abandon it. 



Salet in his admirable work on the Spectra of the Metal- 

 loids, was driven to the conclusion that many of these bodies 

 must be held to possess two spectra. 1 



Although, however, in the views I have expressed I have, 

 as it will be seen, simply followed in the footsteps of such men 



1 His conclusions are thus expressed : 



" Nous avons compare le spectre d'absorption du brome et de 1'iode a leur 

 spectre electrique, et cette comparaison nous semble mettre hors de doute la 

 possibilite des spectres doubles. . . . 



" Nous avons obtenu, par voie electrique, un spectre primaire de 1'iode cor- 

 respondant a son spectre d'absorption. 



" Le soufre, le selenium et le tellure nous ont offert des spectres de combustion 

 tres-analogues aux spectres primaires obtenus par voie electrique, mais different 

 essentiellement des spectres de lignes. . . . 



" Nous avons produit le spectre primaire de 1' azote avec differents corps qui 

 n'ont absolument de commun que 1'azote ; nous pensons done avoir demontre 

 qu'il appartient bien reellement a ce metalloide." Annales de Chemie et de 

 Physique, 4 serie, tome xxviii. pp. 70, 71, 1873. 



O 



